1. Field of the Invention
The present invention relates to a multipath transmission system for transferring signals between multiplex nodes connected to a plurality of networks, and more particularly, to a multipath transmission system for system troubleshooting.
2. Description of the Related Art
Some of multipath transmission systems of this type are used in a system which includes two networks with different response characteristics (e.g., transmission speed). In these first and second networks, a plurality of multiplex nodes transmit signals, such as data, to each other through multiplex buses, thereby effecting independent control.
One of the multiplex nodes constitutes an element common to both networks. This common node has a function as a gateway for relaying signals used in common in the two networks, as well as a communication facility for signal transmission. This multiplex node will hereinafter be referred to as gateway node.
A specific multiplex node in each network periodically transmits signals to the network to which it belongs. This specific node checks reception response signals from the individual multiplex nodes. Alternatively, the specific multiplex node checks to see if the other multiplex nodes are normally operating by periodically transmitting signals to the nodes.
In case of a failure in one of the multiplex nodes, the specific multiplex node infers an occurrence of some trouble from the absence of response signals, transmission failure, etc. if the defective node is in the same network as the specific multiplex node. If the defective node is in the other network to which the specific multiplex node does not belong, the specific node cannot detect it at all. In this multipath transmission system, the control method for each node must be changed, so that the other network should be informed of the details of the failure.
The following is a description of the operation of the system which is supposed to be performed when a certain multiplex node in the first network is disabled from communication due to a failure. First, the specific multiplex node in the first network detects the disabled state of the defective multiplex node by periodic inspection, and then immediately transmits a signal indicative of the occurrence of the failure in the defective node to the first network. On receiving this failure signal, the gateway node relays it to each multiplex node in the second network. Thus, the control method for each multiplex node in the second network can be also changed. The multiplex nodes in the first network can also utilize the failure signal, and the failure, having occurred in the same network, can be detected by any other suitable means. The operation of the system can be supposed to be performed in like manner when the defective multiplex node is in the second network. More specifically, in this case, the specific multiplex node in the second network detects the failure, and the gateway node relays a failure signal to the first network.
The specific multiplex node comprises an emergency communication device and a troubleshooting device, and periodically checks to see if multiplex transmission lines (hereinafter referred to as multiplex buses) are normal. The emergency communication device can perform communication even if one wire of a twisted pair cable, which constitutes each transmission line, is shorted. The troubleshooting device can perform communication only when both wires of the cable are normal. When the specific multiplex node finds, through the periodic inspection, a failure in the network to which it belongs, it turns on an alarm lamp in a meter, for example, thereby informing an operator of the failure. The specific multiplex node may be designed so that it can remember the occurrence of a failure if the failure is a relatively safe one, and can output failure record information when the system is inspected later in a repair shop or the like.
If the communication facility of the gateway node suffers a failure, in this multipath transmission system, a failure signal cannot be relayed to the other network. This is because a relaying function is given to the very gateway node in trouble. Despite the occurrence of the failure, therefore, the control method for each multiplex node in the other network cannot be properly changed. If any of the multiplex buses suffers a failure, that is, if a connector which connects the multiplex bus and its corresponding multiplex node is disengaged, or if the two wires of the bus is shorted, for example, no failure signal is delivered, so that the control method for each multiplex node in the other network cannot be properly changed, as in the aforesaid case.